COMPREHENSIVE MOLECULAR PROFILING MARKEDLY IMPROVES CLINICAL OUTCOME IN CHILDREN WITH RECURRENT CANCER

Purpose: Pediatric cancers are often difficult to classify and can be complex to treat. To ensure precise diagnostics and identify relevant treatment targets, we implemented comprehensive molecular profiling of consecutive pediatric patients with cancer relapse. We evaluated the clinical impact of extensive molecular profiling by assessing the frequency of identified biological onco-drivers, altered diagnosis, and/or identification of new relevant targeted therapies.Patients and Methods: Forty-six tumor samples (44 fresh-frozen; two formalin-fixed paraffin embedded), two bone marrow aspirates, three cerebrospinal fluid samples, and one archived DNA were obtained from 48 children (0–17 years; median 9.5) with relapsed or refractory cancer, where the disease was rapidly progressing in spite of their current treatment or they had exhausted all treatment options. The samples were analyzed by whole-exome sequencing (WES), RNA sequencing (RNAseq), transcriptome arrays, and SNP arrays. Final reports were available within 3–4 weeks after patient inclusion and included mutation status, a description of copy number alterations, differentially expressed genes, and gene fusions, as well as suggestions for targeted treatment.Results: Of the 48 patients, 33 had actionable findings. The most efficient method for the identification of actionable findings was WES (39%), followed by SNP array (37%). Of note, gene fusions were identified by RNAseq in 21% of the samples. Eleven findings led to clinical intervention, i.e. oncogenetic counseling, targeted treatment, and treatment based on changed diagnosis. Four patients received compassionate use targeted therapy. Six patients experienced direct benefits in the form of stable disease or response.Conclusion: In this intervention study, the implementation of comprehensive genetic diagnostics in children with recurrent cancers markedly improved the identification of targetable biological events and patient outcome.